Shuffling is important

2 files changed, 12 insertions, 2 deletions

diff --git a/extract.sh b/extract.sh
index 77c0ddc..e0cb691 100755
--- a/extract.sh
+++ b/extract.sh
@@ -36,7 +36,17 @@ process() {
 		shuf -n $num_games "data/good-playtak-$size" > "data/smalltak-$size"
 		echo -en "\tGenerating training data... "
 		./pptdb "$size" "data/smalltak-$size" generate
-		echo -e "\n\tDone! Wrote $(wc -l data/training-$size.csv | cut -d\  -f1) samples to data/training-$size.csv."
+		echo -e "\tDone, $(wc -l data/training-$size.csv | cut -d\  -f1) samples."
+		echo -en "\tShuffling... "
+		head -n 1 "data/training-$size.csv" > "data/shuf-$size.csv"
+		tail -n+2 "data/training-$size.csv" | shuf >> "data/shuf-$size.csv"
+		mv "data/shuf-$size.csv" "data/training-$size.csv"
+		echo -en "Done.\n\tCompressing data... "
+		if [ -f "data/training-$size.csv.gz" ]; then
+				rm "data/training-$size.csv.gz"
+		fi
+		gzip "data/training-$size.csv"
+		echo -e "Done!\n\tCompressed version available in data/training-$size.csv.gz"
 }
 
 
diff --git a/src/pptdb.c b/src/pptdb.c
index 7179ab0..e681735 100644
--- a/src/pptdb.c
+++ b/src/pptdb.c
@@ -163,9 +163,9 @@ main(int argc, char **argv) {
 		snprintf(td_fn, 64, "data/training-%d.csv",size);
 		training_fh = fopen(td_fn, "w");
 		if (training_fh == NULL) exit(EXIT_FAILURE);
+
 		// Write header
 		/* fputs("\"Player\",\"White flats\",\"Black flats\",",training_fh); */
-
 		for (int depth = 0; depth < max_depth; depth++) {
 			for (int k = 0; k < size*size; k++) {
 				fprintf(training_fh,"\"Stack %d %d\",",depth,k);